llama_cpp_WebUI FILE/ALL Run Models.bat

@echo off
Title 🦙 Llama.cpp Local Server - GPU + Model Selector + Mobile Access

REM --- PATH SETTINGS ---
SET BASE_DIR=D:\Flie\llama.cpp
SET MODELS_DIR=%BASE_DIR%\models
SET SERVER_EXE=%BASE_DIR%\llama-server.exe

REM --- SERVER SETTINGS ---
SET HOST_IP=0.0.0.0
SET PORT=8080
SET GPU_LAYERS=3
SET CONTEXT_SIZE=114096

echo ============================================
echo   🦙 Llama.cpp Local Server - Model Selector
echo ============================================
echo.
echo Available Models in: %MODELS_DIR%
echo.

REM --- LIST ALL MODELS ---
SETLOCAL ENABLEDELAYEDEXPANSION
SET COUNT=0
for %%f in ("%MODELS_DIR%\*.gguf") do (
    SET /A COUNT+=1
    echo !COUNT!. %%~nxf
    SET "MODEL[!COUNT!]=%%~nxf"
)

echo.
echo --------------------------------------------
echo Type "NO"  → Vision Model (Qwen3-VL-2B)
echo Type "Ga"  → Gamma Model (gemma-3-12b)
echo Type "La"  → Llama Model (Llama-3.2-1B)
echo --------------------------------------------
echo.
set /p choice=Enter model number or type NO/Ga/La:

REM --- IF USER TYPES NO ---
IF /I "%choice%"=="NO" (
    echo.
    echo 🧠 Starting Vision Model: Qwen3-VL-2B-Instruct-Q8_0
    echo --------------------------------------------
    start "" "%SERVER_EXE%" --n-gpu-layers 3 --ctx-size 114096 -m "%MODELS_DIR%\Qwen3-VL-2B-Instruct-Q8_0.gguf" --mmproj "%MODELS_DIR%\mmproj-Qwen3-VL-2B-Instruct-Q8_0.gguf" --host %HOST_IP% --port %PORT%
    timeout /t 3 >nul
    goto :MOBILE_URL
)

REM --- IF USER TYPES GA ---
IF /I "%choice%"=="GA" (
    echo.
    echo 🧠 Starting Gamma Model: gemma-3-12b-it-Q4_K_S
    echo --------------------------------------------
    start "" "%SERVER_EXE%" --n-gpu-layers 3 --ctx-size 114096 -m "%MODELS_DIR%\gamma\gemma-3-12b-it-Q4_K_S.gguf" --mmproj "%MODELS_DIR%\gamma\mmproj-model-f16-12B.gguf" --host %HOST_IP% --port %PORT%
    timeout /t 3 >nul
    goto :MOBILE_URL
)

REM --- IF USER TYPES LA ---
IF /I "%choice%"=="LA" (
    echo.
    echo 🧠 Starting Llama Model: Llama-3.2-1B-Instruct-Q8_0
    echo --------------------------------------------
    start "" "%SERVER_EXE%" --n-gpu-layers 3 --ctx-size 114096 -m "%MODELS_DIR%\ollma\Llama-3.2-1B-Instruct-Q8_0.gguf" --mmproj "%MODELS_DIR%\ollma\mmproj-ultravox-v0_5-llama-3_2-1b-f16.gguf" --host %HOST_IP% --port %PORT%
    timeout /t 3 >nul
    goto :MOBILE_URL
)

REM --- NORMAL MODEL SELECTION PATH ---
IF "%choice%"=="" (
    echo No selection made. Exiting...
    pause
    exit /b
)

SET SELECTED_MODEL=!MODEL[%choice%]!
SET MODEL_PATH="%MODELS_DIR%\%SELECTED_MODEL%"

echo.
echo ✅ Selected model: %SELECTED_MODEL%
echo ---------------------------------------------

echo.
echo 🚀 Starting llama-server with %SELECTED_MODEL% ...
echo.

start "" "%SERVER_EXE%" --n-gpu-layers %GPU_LAYERS% --ctx-size %CONTEXT_SIZE% -m %MODEL_PATH% --host %HOST_IP% --port %PORT%
timeout /t 3 >nul

:MOBILE_URL
REM --- GET LOCAL IP FOR MOBILE ACCESS ---
for /f "tokens=2 delims=:" %%a in ('ipconfig ^| findstr /c:"IPv4 Address"') do set LOCAL_IP=%%a
set LOCAL_IP=%LOCAL_IP: =%

echo.
echo 🌐 Open on this PC:      http://127.0.0.1:%PORT%
echo 📱 Open on your mobile:  http://%LOCAL_IP%:%PORT%
echo.
start "" chrome http://127.0.0.1:%PORT%/
pause